BOUTEF: A Multilingual Corpus for FakeNews in North Africa -- Language as a Weapon

The paper introduces BOUTEF, a large-scale multilingual corpus aimed at analyzing the spread of fake news in North Africa, specifically in Algeria and Tunisia. It includes three components: fake narratives, genuine narratives, and user comments, encompassing multiple languages and dialects, such as MSA, Arabic dialects, French, and English. This resource facilitates empirical research on thematic distributions, sentiment patterns, and engagement dynamics, highlighting the significance of emotional narratives in the virality of misinformation, which is crucial for practitioners working on fake news detection and low-resource language processing.